Political scientists have long argued that legislators believe the public is more conservative than it really isāpotentially shaping policies that donāt align with what voters actually want. But what if that story is incomplete?In this episode, we talk with University of Chicago political scientist Adam Zelizer, who challenges the conventional wisdom. His new research suggests that politicians may not be systematically biased to the right, but rather exhibit something he calls āmidpoint biasā.We unpack why this matters: How do politicians perceive their constituents? Are surveys of public opinion misleading policymakersāor are policymakers just inattentive? And what does this all reveal about the messy relationship between democracy, representation, and what voters actually want?
-------- Ā
53:41
--------
53:41
Do We Understand Members Of The Other Party?
Do Democrats and Republicans really misunderstand each other as much as we think?This week, we dive into a surprising new experiment that puts that idea to the test ā literally. Psychologist and researcher Adam Mastriani created a kind of āpolitical Turing test,ā asking people to write persuasive statements from the perspective of the opposite political party. Then, he tested whether others could tell the real from the fake. The results? Most people couldnāt.We unpack what this means for our understanding of polarization, partisan animosity, and political identity. Is the problem really misunderstanding ā or something deeper? Are partisans more empathetic than we give them credit for? Or are they just really good at writing what they think others want to hear?We also explore the experimentās implications for political science research, theory-building, and the broader sociology of science.
-------- Ā
54:14
--------
54:14
MechaHitler and The Political Bias of AI Chatbots
When you ask ChatGPT or Gemini a question about politics, whose opinions are you really hearing?In this episode, we dive into a provocative new study from political scientist Justin Grimmer and his colleagues, which finds that nearly every major large language modelāfrom ChatGPT to Grokāis perceived by Americans as having a left-leaning bias. But why is that? Is it the training data? The guardrails? The Silicon Valley engineers? Or something deeper about the culture of the internet itself?The hosts grapple with everything from āMecha Hitlerā incidents on Grok to the way terms like āunhousedā sneak into AI-generated textāand what that might mean for students, voters, and future regulation. Should the government step in to ensure āpolitical neutralityā? Will AI reshape how people learn about history or policy? Or are we just projecting our own echo chambers onto machines?
-------- Ā
57:28
--------
57:28
Does The Supreme Court Need Term Limits?
We're taking some time off to regroup over the summer, but weāre not just dusting off this older episode for no reason. When we first released it, we were grappling with what Supreme Court reform might look likeāspecifically, whether we should rethink lifetime appointments and move toward term limits.Now? The stakes feel even higher. In just the last few weeks, weāve seen the Court issue decisions that fundamentally reshape presidential powerāoften in ways that donāt reflect where the broader public seems to be. Once again, the question has come roaring back: should nine unelected justices hold this much sway for life?In this episode, we dive into a proposal thatās gaining more traction: fixed 18-year terms for Supreme Court justices. Itās a reform that President Bidenās commission studiedāand one that could change the balance of power in U.S. politics for decades. Whether youāre a reform skeptic or a true believer, this conversation is more relevant than ever. Letās get into it and weāll be back in two weeks with brand new episodes.
-------- Ā
42:06
--------
42:06
Are We Really Hopelessly Divided?
Before we get into todayās episode, we wanted to let you know this is a re-release as we take some time to regroup over the summer. But weāre not just dusting it off for no reason. If anything, this episode feels even more relevant now than when we first aired it.It raises a big question: Are voters really thinking for themselves? Or are they just reflexively rejecting anything the other side says?In this episode, we dig into that question with new experimental research that challenges the conventional wisdom. Are Americans hopelessly divided? Or are they actually more open-minded than we give them credit forāif we present information in the right way?Weāre re-releasing this one because it gives us a more hopeful, evidence-based look at polarizationāand how the way we talk about politics might be part of the problem. And if youāre exhausted by the headlines, this episode just might change how you see American voters.
With all the noise created by a 24/7 news cycle, it can be hard to really grasp what's going on in politics today. We provide a fresh perspective on the biggest political stories not through opinion and anecdotes, but rigorous scholarship, massive data sets and a deep knowledge of theory. Understand the political science beyond the headlines with Harris School of Public Policy Professors William Howell, Anthony Fowler and Wioletta Dziuda. Our show is part of the University of Chicago Podcast Network.
Australia